Professional Documents
Culture Documents
Econ Analysis
Econ Analysis
ScienceDirect
highlights
Article history: Hydrogen has been proposed as energy vector because it can be produced from renewable
Received 25 July 2022 energy sources and its use produces no carbon dioxide. Moreover, many industrial pro-
Received in revised form cesses (e.g., steel industry, ammonia, and fertilizer industry, etc.) are also adopting
8 May 2023 renewable-produced hydrogen as raw material, replacing traditional fossil sources. In this
Accepted 24 June 2023 context, the storage and transportation of hydrogen using liquid organic hydrogen carriers
Available online 10 July 2023 (LOHC) is interesting to deal with the discontinuous production of renewable energies.
The work addressed the study of the hydrogenation process for three compounds, 3,5-
Keywords: dibenzyl toluene, 2-methyl indole and N-ethylcarbazole, on Ni and Ru-based catalysts. The
Hydrogen reactor consisted of a three-phase slurry stirred tank reactor, which has been scale-up
Energy storage together with the rest of the process equipment units using Aspen Hysys simulation
Hydrogenation software. The hydrogen storage cost has been determined applying an economic evalua-
Economic assessment tion to the overall process. The most important costs of the process are the reactor cost and
Optimization the cost of raw materials (the liquid organic hydrogen carrier). The main design variables,
plant capacity, reactor temperature and pressure, were optimized to reduce the storage
cost. The liquid organic hydrogen carrier with the minimum storage cost at the optimum
operating conditions is 3,5-dibenzyl toluene with a reactor configuration of two parallel
tanks.
© 2023 The Author(s). Published by Elsevier Ltd on behalf of Hydrogen Energy Publications
LLC. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
* Corresponding author.
n
E-mail address: sordonez@uniovi.es (S. Ordo ~ ez).
https://doi.org/10.1016/j.ijhydene.2023.06.273
0360-3199/© 2023 The Author(s). Published by Elsevier Ltd on behalf of Hydrogen Energy Publications LLC. This is an open access article under the CC BY-NC-
ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1114 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 5 2 ( 2 0 2 4 ) 1 1 1 3 e1 1 2 3
Freezing 39 C 58 C 70 C
Boiling 390 C 273 C 295 C
Stoichiometry coef. 9 mol H2/mol LOHC 4 mol H2/mol LOHC 6 mol H2/mol LOHC
Reaction enthalpy 530 kJ/mol 206 kJ/mol 342 kJ/mol
Weight capacity 6.2 wt% 5.8 wt % 5.6 wt%
Volumetric capacity 762 L H2/L 705 L H2/L 736 L H2/L
H2 solubility at 1 MPa, 150 C 17.6 mol/m3 16.3 mol/m3 17 mol/m3
The other two compounds, 2-methyl indole and N-ethyl- catalyst. This reaction is accomplished according to three
carbazole, are both heterocycle compound with equivalent steps in series: (1) hydrogen transfer and dissolution in the
properties. Like for 3,5-dibenzyl toluene, the hydrogenation liquid phase, (2) hydrogen transfer to the catalyst surface and
takes place using supported precious metal catalysts at (3) reaction on the catalyst surface. Using the film theory to
similar conditions [36,37,43e45]. These compounds are also describe the mass transfer steps and assuming a reaction ki-
stable and non-toxic, but the hydrogenated forms have high netics of first order with respect to each reactant (LOHC and
freezing points, which can complicate the storage. The main H2), the following equations can be formulated:
advantage of these compounds is their lower dehydrogena-
tion temperature [34]. rH2 ¼ akL pH2 HH2 CH2 liq (1)
The proposed reactor consisted of a continuous three-phase
stirred tank slurry reactor. Hydrogen is fed in excess, so rH2 ¼ fs aP kC CH2 liq CH2 sup (2)
unreacted hydrogen is separated and recycled to the reactor.
This reactor type is more advantageous for heat and mass rH2
rLOHC ¼ ¼ fs kv CLOHC CH2 sup (3)
transfer. The hydrogenation reactions are very exothermic (see nH2
reaction enthalpies reported in Table 1) and, given that no Where rH2 and rLOHC are, respectively, the hydrogen and
solvents are used, temperature must be carefully controlled. LOHC, reaction rates per unit volume of reactor, nH2 is the
Stirred tanks equipped with a jacked for refrigeration are sim- stoichiometric coefficient of H2 in the hydrogenation reaction
ple and efficient. In stirred tank reactors, the catalyst is in the (see Table 1), akL is the gas-to-liquid mass transfer coefficient,
form of a slurry. To facilitate the suspension, the catalyst par- aP kC is the liquid-to-solid mass transfer coefficient, kv is the
ticles must be of small size, which means that mass transfer to kinetic constant per unit volume of catalyst (i.e., kv ¼ kw rcat ),
the particles and in the inside is fast. The use of a stirred also pH2 is the partial pressure of hydrogen in the gas phase, HH2 is
helps to transfer and dissolve hydrogen in the liquid phase. For the Henry's law constant for hydrogen, CH2 liq and CH2 sup are,
these reasons, the use of stirred tank reactors is a good option respectively, the molar hydrogen concentrations in the bulk
to carry out the hydrogenation process at the industrial scale. liquid and the surface of the catalyst, CLOHC is the molar con-
The reactor scale-up has been based on a kinetic model of centration of LOHC, and fs is the volumetric fraction of catalyst
the respective hydrogenation catalytic reactions; the influ- in the reactor.
ence of mass transfer on reaction rate has been estimated and Equations (1)e(3) have been solved for CH2 liq and CH2 sup , to
discussed. In the economic evaluation section, the capital obtain an expression of the reaction rate as a function of
investment and operating costs of the process are estimated hydrogen partial pressure, the transport coefficients and the
and used to determine the unitary cost of hydrogen storage. kinetic constant. The reaction rate of the LOHC is given by the
Finally, the main operating and design variables of the process following expression, where each term of the denominator
are optimized by means of a sensitivity analysis with the aim corresponds to a resistance (respectively, mass transfer to the
of decreasing the unitary cost of hydrogen storage. liquid phase, mass transfer to the catalyst surface and
reaction):
.
Material and methods pH2 HH2
rH2
rLOHC ¼ ¼ nH2 n (4)
nH2 akL
þ fs aHP2kC þ fs kv C1LOHC
Kinetic model
For the case of fast mass transfer steps with respect to the
The catalytic hydrogenation of the LOHC is a three-phase re- reaction, eq. (4) can be simplified eliminating the first two
action involving gaseous hydrogen, liquid LOHC and the solid terms of the denominator:
1116 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 5 2 ( 2 0 2 4 ) 1 1 1 3 e1 1 2 3
.
rLOHC ¼ fs kv CLOHC pH2 HH2 (5) Results and discussion
The reactions considered in the present work are the hy-
Assessment of mass transfer limitations
drogenation of 3,5-dibenzyl toluene on Ni Raney, and 2-
methyl indole and N-ethylcarbazole on Ru/Al2O3. The kinetic
The reactor selected to carry out the three-phase hydrogena-
constants have been calculated using reaction experiments
tion reaction consists of a slurry stirred tank. This reactor is
from the bibliography, which were carried out in the absence
equipped with a mechanical stirrer to maintain the solid
of mass transfer resistances [41,44,45]. Arrhenius-type
catalyst in suspension and a diffuser to disperse hydrogen gas
dependence with temperature of the kinetic constant is
in the liquid LOHC. In order to help the suspension of the
assumed, kv ¼ kv0 eEa =RT , where kv0 is the pre-exponential
catalyst, a particle size of 0.1 mm and a fraction catalyst of
factor, Ea is the activation energy, R is the ideal gas constant
0.5 vol % have been selected [47]. As shown in eqs. (4) and (5),
and T is temperature. Table 2 shows the fitted parameters of
higher catalyst fractions (fw ) result in higher reaction rates but
the kinetic model, eq. (5).
hindering the catalyst-liquid mixing. In these multi-phase
systems, good mixing inside the reactor is ensured using, at
Process simulation
least, a stirring speed of 90 rpm and a specific power of 1.6 kW/
m3 [47].
The study of the hydrogenation process has been carried out
The mass transfer coefficients of eq. (4) have been deter-
by simulating with the software Aspen Hysys. Peng-Robinson
mined using correlations for stirred tanks [48,49]. These cor-
thermodynamic model was selected, and the properties of the
relations have been evaluated using the above parameters
hydrogenated compounds were introduced in the model.
and the physical properties, estimated at a temperature of
Fig. 1 shows the flowsheet of the hydrogenation process.
190 C. The results of the calculations are summarized in
The hydrogenation reactor is a three-phase slurry reactor,
Table 3. As shown, the estimated mass transfer resistances
equipped with a mechanical stirrer to maintain the solid
are much lower than the reaction one. This means that the
catalyst in suspension and a diffuser to disperse hydrogen gas
reaction step is the slowest one and controls the overall re-
in the liquid LOHC. This reactor has been modelled using the
action rate. The main operating and design variables affecting
continuous stirred tank reactor model, which assumes perfect
the mass transfer resistances are stirring speed (90 rpm) and
mixture inside the reactor. The kinetic model of section 2.1
power (1.6 kW/m3), catalyst particle size (0.1 mm) and catalyst
has been used to predict reaction rates. Two cases have been
fraction (0.5 vol %). The values considered for these variables
considered, a reactor made of one tank and two tanks in se-
were chosen to maintain good catalyst suspension and mix-
ries; the latter one corresponds to the flowsheet shown in
ing, and, at the same time, they were enough to make mass
Fig. 1.
transfer resistances negligible.
Hydrogen fed to the process is produced in an electrolytic
cell using renewable electricity and water. Typical electrolytic
Process scale-up and simulation
cells operate at 6 MPa and 80 C [46], so these are the conditions
chosen for the hydrogen feed stream. The reactor operates
The hydrogenation process has been simulated and scaled up
using hydrogen in excess to shift the reaction towards products.
considering the kinetic model presented in section 2.2. The
However, given its low solubility, not all the hydrogen is dis-
nominal capacity supposed for the process is 6 MW of
solved in the liquid. As shown in Fig. 1, the hydrogen streams
hydrogen power, which corresponds to a H2 consumption of
leaving the reactors without dissolving and reacting are recy-
1578 t/year. According to the reaction kinetics, a reactor
cled and mixed with the fresh feed. This loop of hydrogen is
temperature of 170 C and pressure of 6 MPa are appropriate to
important to maintain the required excess of hydrogen and a
operate the reactor, so these values were fixed as a first esti-
good dispersion of the gas inside the liquid phase.
mation. These variables will be optimized in the sensitivity
The LOHC is introduced in the tanks using pumps. At the
analysis of section 3.4.
end of the process, the hydrogenated form of the LOHC is
Considering these specifications, the reactor volume
cooled in a heat exchanger and expanded to atmospheric
required to achieve 99% conversion of the LOHC was calcu-
pressure. This causes that a bit of dissolved unreacted
lated. A height of liquid inside the tank equals to the diameter
hydrogen desorbs from the liquid phase. This gas can be
is assumed to size the tank. For the case of 3,5-dibenzyl
vented or, otherwise, collected, recompressed, and recycled to
toluene and 2-methyl indole, it is considered the possibility
the reactor.
of using either a reactor made of one tank or, otherwise, two
tanks in series. The latter option is interesting to reduce the
Table 2 e Hydrogenation kinetic model of the LOHC used total reactor volume for the same final conversion, because a
in the present work [41,44,45]. higher LOHC concentration in the first tank increases the re-
action rate. For N-ethylcarbazole, the reaction rate is lower
3,5-dibenzyl 2-methyl N-
toluene indole ethylcarbazole and the option of using only one tank is discharged, because of
the large volume required.
Catalyst Ni-Raney Ru/Al2O3 Ru/Al2O3
The hydrogenation reactions are exothermic (the en-
ln kv0 (m3/mol s) 12.1 18.2 22.5
Ea (kJ/mol) 48.4 73.8 93.9 thalpies of the reaction are displayed in Table 1), so the re-
Temperature (ºC) 150e200 120e170 120e170 actors require refrigeration to maintain the desired reaction
Source data [41] [44] [45] temperature. The refrigeration of the reactors is done using
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 5 2 ( 2 0 2 4 ) 1 1 1 3 e1 1 2 3 1117
Economic evaluation
Table 3 e Assessment of mass transfer limitations in the
slurry reactors at 190 C.
The aim of this section is to carry out an estimation of the
3,5- 2- N- costs of the process, in order to calculate the cost of H2 stor-
dibenzyl methyl ethylcarbazole age. The cost of the process is divided into capital investment
toluene indole
and operating costs.
Kinetic constant, kv 0.6 0.2 0.01 The capital investment has been estimated as a function of
(m3/mol s)
the cost of the main pieces of equipment (reactors, separator,
Gas-to-liquid mass transfer 0.096 0.12 0.10
compressor, etc.). The purchase cost of the equipment units
coefficient, akL (s1) [48]
Liquid-to-solid catalyst 215 255 232 has been calculated using correlations from the bibliography
mass transfer coefficient, [50], which depends of the corresponding design variables
aP kC (s1) [49] displayed in Table 4. The purchase costs have been updated to
Resistances (s) 2021 using the Chemical Engineering Process Cost Index
Gas-to-liquid, 1= akL 10.4 8.3 10 (CEPCI). The delivery cost of the equipment units is estimated
Liquid-to-solid, 1= fs aP kC 0.93 0.78 0.86
as 10% of the purchase cost. The capital investment includes
Reaction, 1= fs kv CLOHC 7.1 $ 104 21.3 $ 104 426 $ 104
the fixed capital (cost of equipment, installation, piping,
instrumentation, electrical, building, services, engineering,
cooling water that flows through the reactor jacket, given that contingencies, etc.) and working capital (estimated as 15% of
the use of coils is not recommended for slurry reactors. The the total investment). The total capital investment is calcu-
area needed for the reactor jacket is calculated using an lated as 6.05 times the total cost of the delivered equipment
overall heat transfer coefficient of 0.5 kW/m2 K. In case this units. Further details can be found in the bibliography [51].
area is larger than the available in the tank, its geometry is The results of these estimations are summarized in Table 5.
adjusted to fulfil with the cooling requirements. Table 4 shows The LOHC process requiring the lowest capital investment
the results of the scale-up for the main units of equipment. is the one based on 3,5-dibenzyl toluene and a reactor of one
The power consumed by the impeller of the reactor, the tank. The options based on a two-tank reactor require, in
compressor and pumps has also been calculated. This is general, a higher capital investment. The piece of equipment
needed to estimate the consumption of electricity and the with the higher contribution to the total cost is the reactor
operating costs. (40e50%), so the cost of an additional tank has an important
influence. The process based on N-ethyl carbazole requires
the higher capital investment.
Table 4 e Scale-up of the main units of equipment.
The total investment has been annualized considering a
3,5- 2-methyl N-ethyl 10-year period and an interest rate of 15%; the corresponding
dibenzyl indole carbazole annual capital charge ratio (ACCR) is 0.199 [50].
toluene
The operating costs are divided into fixed cost (insurances
1 2 1 2 2 and taxes) and variable costs (raw materials, labour, utilities,
tank tanks tank tanks tanks etc.). The estimates of these costs according to the method-
H2 feed (kg/h) 185 186 180 190 182 ology proposed in the bibliography [51] is summarized in
LOHC flow rate (m3/h) 3.35 3.35 3.28 3.42 3.35 Table 6. The main costs correspond to raw materials, utilities
Reactor size (m3, each 8 8 8 6 6
(cooling water, 2.17 V/1000 kg) and electricity (0.0645 V/kWh).
tank)
Regarding the raw materials, it is considered that most of
Compressor (kW) 21.6 23.7 23 20 115
Pump (kW) 7.5 7.5 7.4 7.7 7.6 the LOHC can be re-used after H2 is released in the dehydro-
Heat exchanger area 12 13 13 13 12 genation process. For estimating purposes, 90% of the LOHC
(m2) feed flow rate would correspond to re-used compound and the
Separator (m3) 0.31 0.31 2.6 0.43 1.52 remaining 10% to make-up. The make-up would compensate
1118 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 5 2 ( 2 0 2 4 ) 1 1 1 3 e1 1 2 3
losses of LOHC in the transportation, storage, or hydrogena- dibenzyl toluene and 2-methyl indole are remarkably close, in
tion/dehydrogenation processes. Therefore, only the cost of the range 3.22e3.45 V/kg. The cost of N-ethyl carbazole is
that 10% of LOHC is accounted for in the operating costs of clearly higher 5.38 V/kg.
Table 6. The prices of the LOHC are 0.90 V/kg for 3,5-dibenzyl
toluene, 0.86 V/kg for 2-methyl indole, and 1.64 V/kg for N- Sensitivity analysis
ethyl carbazole. It should be noted that as the storage tech-
nology is spread and optimized, the LOHC recovery may in- The aim of the sensitivity analysis is the study of the main
crease and the amount of make-up considerably reduced, operation and design variables on the production cost of
improving the economy of the process. hydrogen. The considered variables are plant capacity
The operating costs are quite similar for 3,5-dibenzyl (measured in terms of hydrogen power), reactor temperature
toluene and 2-methyl indole for the different process config- and pressure. These variables have been varied one at a time
urations with one and two tanks. However, they are consid- with respect to the base case of sections 3.2 and 3.3 (respec-
erably higher for N-ethyl carbazole, which is due to the higher tively, 6 MW of hydrogen power, 170 C and 6 MPa). For each
price of this LOHC. combination of variables, the main pieces of equipment of the
Table 7 shows a summary of the total annual costs, plant (i.e., reactor, vessels, heat exchangers and pumping)
including the annualized investment and the operating costs, have been sized, the capital and operational costs determined,
for the different alternative LOHC. The lowest total annual and the production cost of hydrogen calculated, as explained
cost corresponds to 3,5-dibenzyl toluene and the case of using in sections 3.2 and 3.3.
a reactor of one tank. The cost of H2 storage has also been Fig. 2 (left) shows the results for the process based on 3,5-
calculated using the actual plant capacity and H2 consump- dibenzyl toluene as LOHC. The configuration of a reactor
tion. The nominal design H2 plant capacity was fixed to 6 MW. made of one and two tanks has been compared in terms of
However, after the scale-up, the actual plant capacity can be the production cost of hydrogen. The plant capacity has the
slightly higher (deviations are within 5%). This is mainly due highest influence on the production cost of hydrogen, which
to reactor over-sizing to ensure an adequate heat removal to decreases sharply in the range 2e10 MW; above 10 MW, with
the tank jacket. H2 costs around 3 V/kg, the influence is less marked. The
For 3,5-dibenzyl toluene, the cost of H2 storage is 3.22 V/kg curves for one and two tanks are close, being the configura-
for the one-tank configuration. Nevertheless, the costs for 3,5- tion made of one tank more favourable. This curve is
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 5 2 ( 2 0 2 4 ) 1 1 1 3 e1 1 2 3 1119
Table 7 e Summary of process costs and calculation of H2 storage cost (currency in Euro).
3,5-dibenzyl toluene 2-methyl indole N-ethyl carbazole
1 tank 2 tanks 1 tank 2 tanks 2 tanks
ANNUALIZED INVESTMENT 323 000 579 000 392 000 516 000 648 000
OPERATING COST 4 881 900 5041 000 5 000 500 5017 800 7 955 100
TOTAL ANNUAL COST 5 204 900 5620 000 5 392 500 5533 800 8 603 100
H2 plant (MW) 6.2 6.2 6.0 6.3 6.1
H2 consumption (t H2/year) 1630 1630 1580 1660 1600
Storage cost (V/kg H2) 3.22 3.45 3.41 3.32 5.38
Fig. 2 e Sensitivity analysis of the main design variables on hydrogen cost. Left graphs: 3,5-dibenzyl toluene with a reactor
of one ( ) and two ( ) tanks in series. Right graphs: 2-methyl indole with a reactor of one ( ) and two ( ) tanks in series.
1120 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 5 2 ( 2 0 2 4 ) 1 1 1 3 e1 1 2 3
Table 8 e Summary of the economic analysis for the optimum conditions: 18 MW of plant capacity, 190 C of reactor
temperature and 6 MPa of pressure (currency in Euro).
3,5-dibenzyl toluene 2-methyl indole
2 tanks parallel 2 tanks series 1 tank 2 tanks series
Reactor size (m3, each tank) 14 40 25 25
TOTAL CAPITAL INVESTMENT 3 700 000 6550 000 3 400 000 5070 000
ANNUALIZED INVESTMENT 740 000 1300 000 670 000 1 010 000
OPERATING COSTS 12 540 000 12 950 000 12 710 000 12 550 000
TOTAL ANNUAL COST 13 280 000 14 250 000 13 380 000 13560 000
H2 plant power (MW) 18.5 18.6 17.8 18.1
H2 consumption (t H2/year) 4860 4890 4680 4760
H2 storage cost (V/kg H2) 2.73 2.92 2.86 2.86
Process scale-up and economic analysis at optimum production with periods of surplus (for example, in spring-
conditions summer) and periods of depletion, when the plant must be
stopped. The economic analysis has been carry out consid-
The plant has been scale-up at the optimum conditions ering a 10-year period (with interest rate of 15%) to annualise
determined in the sensitivity analysis of the previous section: the capital investment and that the plant operates for 300 days
plant capacity of 18 MW, reactor temperature of 190 C and every year. In case the plant operated for fewer days a year,
pressure of 6 MPa. Table 8 summarizes the main economic the consumption of H2 and generation of LOHC would
indicators and the H2 storage cost for the two LOHC that decrease and, hence, the cost of H2 storage would increase. To
performed better: 3,5-dibenzyl toluene and 2-methyl indole. evaluate the impact of this, one should examine the contri-
The reactor configurations with one and two tanks were bution of the different costs to the total annual costs. Thus,
considered. However, for 3,5-dibenzyl toluene, the use of just 94% of the total annual costs correspond to operating costs
one single tank would require large tank size, exceeding the (Table 8), of which 74% are due to raw materials (mainly LOHC
maximum recommended. Therefore, the use of two tanks in make-up) and utilities (Table 6). When the plant stops, raw
parallel of equal size was considered for this case. materials and utilities are not consumed. Since they are the
The economic analysis for the optimum conditions in- greater contributors to the total annual costs (70% of them)
dicates that the process with the lowest capital investment is and the H2 storage cost, the impact of fewer operating days a
the one based on 2-methyl indole with a reactor of one tank. year would be limited.
The operating costs are quite similar for all the processes, the To verify this, the influence on the H2 storage cost of the
lowest one being that based on 3,5-dibenzyl toluene with a annual usage of the plant capacity has been evaluated for the
reactor of two tanks in parallel. This process is also the one case of the 3,5-dibenzyl toluene process with a reactor of two
with the lowest total annual cost and hydrogen storage cost, tanks in parallel (the process with the lowest H2 storage cost).
2.73 V/kg. Consequently, this is the recommended process The nominal plant capacity is 18 MW, (Table 8). As shown in
and operating conditions of all the processes studied in this Fig. 4a, if the annual use of plant capacity decreases to a 50% of
work. The use of a reactor made of two tanks in parallel reduce the nominal, the storage cost will increase ca. 20%.
considerably the capital investment and makes it easier the In the calculations, it was assumed a LOHC recovery of 90%
control of the reactor temperature. considering all the stages, i.e., hydrogenation, storage, and
The generation of hydrogen using renewable energy (e.g., dehydrogenation. The LOHC recovery is negatively affected by
solar and wind power) may suffer from intermittent undesired reactions, taking place during the hydrogenation
Fig. 4 e Analysis of the influence of the annual use of plant capacity (a) and LOHC recovery yield (b) on the hydrogen storage
cost. Nominal plant capacity: 18 MW. LOHC: 3,5-dibenzyl toluene with a reactor of two tanks in parallel. Bar diagrams:
annualized investment and operating costs.
1122 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 5 2 ( 2 0 2 4 ) 1 1 1 3 e1 1 2 3
[20] Makepeace JW, He T, Weidenthaler C, Jensen TR, Chang F, [35] Oh J, Jeong K, Kim TW, Kwon H, Han JW, Park JH, et al. 2-(N-
Vegge T, et al. Reversible ammonia-based and liquid organic Methylbenzyl)pyridine: a potential liquid organic hydrogen
hydrogen carriers for high-density hydrogen storage: recent carrier with fast H2 release and stable activity in consecutive
progress. Int J Hydrogen Energy 2019;44:7746e67. cycles. ChemSusChem 2018;11:661e5.
[21] Ferreira APRA, Oliveira RCP, Mateus MM, Santos DMF. A [36] Yang X, Wu Y, Yu H, Sun M, Zheng J, Li X, et al. A YH3
review of the use of electrolytic cells for energy and promoted palladium catalyst for reversible hydrogen
environmental applications. Energies 2023;16:1593. storage of N-ethylcarbazole. Int J Hydrogen Energy
[22] Bollmann J, Schmidt N, Beck D, Preuster P, Zigan L, 2020;45:33657e62.
Wasserscheid P, et al. A path to a dynamic hydrogen storage [37] Fei S, Han B, Li L, Mei P, Zhu T, Yang M, et al. A study on the
system using a liquid organic hydrogen carrier (LOHC): catalytic hydrogenation of N-ethylcarbazole on the
burner-based direct heating of the dehydrogenation unit. Int mesoporous Pd/MoO3 catalyst. Int J Hydrogen Energy
J Hydrogen Energy 2023;48:1011e23. 2017;42:25942e50.
[23] Kim C, Lee Y, Kim K, Lee U. Implementation of formic acid as [38] Yang M, Dong Y, Fei S, Ke H, Cheng H. A comparative study of
a liquid organic hydrogen carrier (LOHC): techno-economic catalytic dehydrogenation of perhydro-N-ethylcarbazole
analysis and life cycle assessment of formic acid produced over noble metal catalysts. Int J Hydrogen Energy
via CO2 utilization. Catalysts 2022;12:1113. 2014;39:18976e83.
[24] Hamayun MH, Hussain M, Maafa IM, Aslam R. Integration of [39] Gianotti E, Taillades-Jacquin M, Rozie re J, Jones DJ. High-
hydrogenation and dehydrogenation system for hydrogen purity hydrogen generation via dehydrogenation of organic
storage and electricity generation e simulation study. Int J carriers: a review on the catalytic process. ACS Catal
Hydrogen Energy 2019;44:20213e22. 2018;8:4660e80.
[25] Preuster P, Papp C, Wasserscheid P. Liquid organic hydrogen [40] Brückner N, Obesser K, Bo € smann A, Teichmann D, Arlt W,
carriers (LOHCs): toward a hydrogen-free hydrogen Dungs J, et al. Evaluation of industrially applied heat-transfer
economy. Acc Chem Res 2017;50:74e85. fluids as liquid organic hydrogen carrier systems.
[26] Niermann M, Beckendorff A, Kaltschmitt M, Bonhoff K. ChemSusChem 2014;7:229e35.
Liquid organic hydrogen carrier (LOHC) e assessment based [41] Ali A, Kumar GU, Lee HJ. Investigation of hydrogenation of
on chemical and economic properties. Int J Hydrogen Energy Dibenzyltoluene as liquid organic hydrogen carrier. Mater
2019;44:6631e54. Today Proc 2021;45:1123e7.
[27] Lee S, Kim T, Han G, Kang S, Yoo Y-S, Jeon S-Y, et al. [42] Shi L, Qi S, Qu J, Che T, Yi C, Yang B. Integration of
Comparative energetic studies on liquid organic hydrogen hydrogenation and dehydrogenation based on
carrier: a net energy analysis. Renew Sustain Energy Rev dibenzyltoluene as liquid organic hydrogen energy carrier.
2021;150:111447. Int J Hydrogen Energy 2019;44:5345e54.
[28] Markiewicz M, Zhang Y-Q, Empl MT, Lykaki M, Tho € ming J, [43] Wan C, An Y, Xu G, Kong W. Study of catalytic hydrogenation
Steinberg P, et al. Hazard assessment of quinaldine-, of N-ethylcarbazole over ruthenium catalyst. Int J Hydrogen
alkylcarbazole-, benzene- and toluene-based liquid organic Energy 2012;37:13092e6.
hydrogen carrier (LOHCs) systems. Energy Environ Sci [44] Li L, Yang M, Dong Y, Mei P, Cheng H. Hydrogen storage and
2019;12:366e83. release from a new promising liquid organic hydrogen
[29] Dı́az E, Rapado-Gallego P, Ordon~ ez S. Systematic evaluation storage carrier (LOHC): 2-methylindole. Int J Hydrogen
of physicochemical properties for the selection of alternative Energy 2016;41:16129e34.
liquid organic hydrogen carriers. J Energy Storage [45] Sotoodeh F, Zhao L, Smith KJ. Kinetics of H2 recovery from
2023;59:106511. dodecahydro-N-ethylcarbazole over a supported Pd catalyst.
[30] Hiyoshi N, Rode CV, Sato O, Shirai M. Biphenyl Appl Catal, A 2009;362:155e62.
hydrogenation over supported transition metal catalysts [46] Runge P, So € lch C, Albert J, Wasserscheid P, Zo € ttl G,
under supercritical carbon dioxide solvent. Appl Catal, A Grimm V. Economic comparison of different electric fuels
2005;288:43e7. for energy scenarios in 2035. Appl Energy 2019;233e234:
[31] Leinweber A, Müller K. Hydrogenation of the liquid organic 1078e93.
hydrogen carrier compound monobenzyl toluene: reaction [47] Woods DR. Rules of thumb in engineering practice. Wiley-
pathway and kinetic effects. Energy Technol 2018;6:513e20. VCH; 2007.
[32] Huynh N-D, Hur SH, Kang SG. Tuning the dehydrogenation [48] Azzopardi BJ, Mudde RF, Lo S, Morvan H, Yan Y, Zhao D.
performance of dibenzyl toluene as liquid organic hydrogen Sparged stirred vessels. Hydrodynamics of gas-liquid
carriers. Int J Hydrogen Energy 2021;46:34788e96. reactors. 2011. p. 61e89.
[33] Hu P, Fogler E, Diskin-Posner Y, Iron MA, Milstein D. A novel [49] Biardi G, Baldi G. Three-phase catalytic reactors. Catal Today
liquid organic hydrogen carrier system based on catalytic 1999;52:223e34.
peptide formation and hydrogenation. Nat Commun [50] Towler G, Sinnott R. Chemical engineering design. Principles,
2015;6:6859. practice and economics of plant and process design. 2nd ed.
B, Byun M, Lim H. Design, economic evaluation,
[34] Brigljevic Elsevier; 2008.
and market uncertainty analysis of LOHC-based, CO2 free, [51] Peters MS, Timmerhaus KD, West RE. Plant design and
hydrogen delivery systems. Appl Energy 2020;274:115314. economics for chemical engineers. 5º ed. McGraw-Hill; 2003.